-
Posts
31 -
Joined
-
Last visited
Recent Profile Visitors
The recent visitors block is disabled and is not being shown to other users.
-
[SOLVED] 100% /var/log usage & rsyslogd high CPU load
ioncube replied to ioncube's topic in Radxa Zero 3W/3E
Solution found at -
My OS OS: Armbian (24.8.4) aarch64 Host: Radxa ZERO 3 Kernel: 6.1.75-vendor-rk35xx Uptime: 5 mins Packages: 1358 (dpkg) Shell: bash 5.2.21 Terminal: /dev/pts/0 CPU: (4) @ 2.000GHz Memory: 1168MiB / 7686Mi Reference Posts x2 I have this patch already in my /etc/systemd/journald.conf I edited /etc/default/armbian-zram-config -> ZRAM_PERCENTAGE=10 but allotment size didn't increase after a reboot (~47M before as well). By default this variable was commented out Filesystem Size Used Avail Use% Mounted on tmpfs 769M 24M 745M 4% /run /dev/mmcblk1p2 116G 32G 84G 28% / tmpfs 3.8G 100K 3.8G 1% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 3.8G 4.0K 3.8G 1% /tmp /dev/md127 439G 65G 352G 16% /media/radxaRaid480gb /dev/mmcblk1p1 256M 157M 100M 62% /boot /dev/sdd1 117G 37G 75G 34% /media/radxa128gb /dev/zram1 47M 46M 0 100% /var/log
-
OrangePi 4A Bluetooth Mouse Keyboard connectivity issue Solved The solution to the problem when mouse pairs / connects with BT but doesnt work Prebuilt kernel files are available https://github.com/defencedog/orangepi4A/blob/main/Bluetooth%20Mouse%20Not%20Working%20Solved.md
-
I have having the same issue as discussed here & it also provided solution for it https://github.com/IceWhaleTech/CasaOS/issues/1136 I have added cgroup_enable=cpuset cgroup_enable=memory cgroup_memory=1 to armbianEnv.txt The dmesg logs indicate that these arguments are failing [ 15.630904] Kernel command line: root=UUID=eb6a27ef-3932-431f-92fd-e61fea75356c rootwait rootfstype=ext4 splash=verbose console=ttyS2,1500000 console=tty1 consoleblank=0 loglevel=1 ubootpart=f3aeb909-d11b-ec42-a52c-4507e474335c usb-storage.quirks=0x2537:0x1066:u,0x2537:0x1068:u,0x1058:0x25a2:u cma=256M cgroup_enable=cpuset cgroup_enable=memory cgroup_memory=1 cgroup_enable=cpuset cgroup_memory=1 cgroup_enable=memory [ 15.631910] Unknown kernel command line parameters "splash=verbose ubootpart=f3aeb909-d11b-ec42-a52c-4507e474335c cgroup_enable=memory cgroup_memory=1", will be passed to user space. . . . [ 18.261976] Run /init as init process [ 18.262019] with arguments: [ 18.262028] /init [ 18.262034] with environment: [ 18.262039] HOME=/ [ 18.262044] TERM=linux [ 18.262049] splash=verbose [ 18.262054] ubootpart=f3aeb909-d11b-ec42-a52c-4507e474335c [ 18.262060] cgroup_enable=memory [ 18.262065] cgroup_memory=1 [ 18.263469] mmcblk1: p1 p2 The bug still haunts me, any ideas how should I approach this problem
-
@Fajar Muttaqin Can you share your compiled kernel & headers `.debs` In default kernel support for filesystems `cifs` & `nfs` were not included. I compiled with these support as well https://github.com/defencedog/orangepi4A/tree/main/kernels
-
@Angel Luis Pérez Can you please upload your debian image. Plus kindly explain which hardware part is not working?
-
Anyone got a working image? Any review of what is working & what is not
-
Efforts to develop firmware for H96 MAX V56 RK3566 8G/64G
ioncube replied to Hqnicolas's topic in Rockchip CPU Boxes
@Hqnicolas [ 6.519079] rk817-codec rk817-codec: rk817_probe: chip_name:0x80, chip_ver:0x94 [ 6.525779] input: rockchip-rk809 Headset as /devices/platform/rk809-sound/sound/card0/input2 I am using OPi3b v2.1 with same RK817 audio, however I am failing for past month & my 3.5mm audio out is not working under vendor 6.1x kernel (both from Armbian / Joshua) interestingly both have same error! However mainline kernel has no problem but because of H265 VPU I prefer to use vendor kernel. Can you please help. My fiddlings are mentioned ar github https://github.com/Joshua-Riek/ubuntu-rockchip/issues/1041 -
@Giordano Sologuren here its a stable OS only 1xUSB2.port & 3.5mm audio out is not working but I am working on it reddit github link For the latest Armbian_community_24.11.0-trunk.66_Orangepi3b_noble_vendor_6.1.75_gnome-kisak_desktop.img.xz nothing works. Boot doesn't occur putty.log
-
orangepi3b initial startup is ok; wifi fails after reboot
ioncube replied to Alexander Buzin's topic in Rockchip
@Kutia which Opi3b you are using. For v2.1 I have Broadcom AP6256 ... wifi I was able to work but BT no ...I am looking answers for latter -
@Protx I will test your images. But tell me have you ever made BT working? Opi3b v2.1 uses AP6256 Broadcom..
-
Struggle continues... Because the hdmi is not working (error logs mentioned above) I tried uninstalling armbian kernels & installed official vendor .deb files, These I compiled using OPi official build procedure for OPi 3b v2.1 You can get x3 .debs from .7z archive here https://github.com/defencedog/orangepi3b_v2.1/tree/b0d58e10589c567025774da07ea300e48631740f/Orangepi-3B%20rk3566-5.10.160-panfrost/cross-built then apt list --installed |grep -e ^linux-image -e ^linux-dtb -e ^linux-headers sudo apt purge linux-image-xxx linux-dtb-xxx linux-headers-xxx sudo apt install libssl-dev python-is-python3 sudo dpkg -i linux-dtb-legacy-rockchip-rk356x_1.0.6_arm64.deb linux-headers-legacy-rockchip-rk356x_1.0.6_arm64.deb linux-image-legacy-rockchip-rk356x_1.0.6_arm64.deb sudo sync because of changed dtb & overlays modify /boot/armbianEnv.txt Following x2 lines must be same. This reflect changes in /boot/dtb/rockchip/ & /boot/dtb/rockchip/overlays overlay_prefix=rk356x fdtfile=rockchip/rk3566-orangepi-3b-v2.dtb Save edits & reboot Success HDMI now works. BT doesn't work (no module found). WIFI also works however Broadcom now uses wl driver instead of brcmfmac I installed ubuntu-desktop & the cursor is invisible, however cursor works OK while logged in as X-session of Ubuntu meaning wayland is to blame. If you logged in KODI wayland session cursor works OK?? GDM3 has no cursor as well When I move cursor while in wayland session, these errors keep repeating journalctl -f Aug 26 20:41:02 orangepi3b gnome-shell[2921]: Failed to post KMS update: drmModeAtomicCommit: Invalid argument Aug 26 20:41:02 orangepi3b kernel: [drm:vop2_plane_atomic_check] *ERROR* Unsupported linear format at Cluster0-win0 Aug 26 20:41:02 orangepi3b gnome-shell[2921]: Page flip discarded: drmModeAtomicCommit: Invalid argument Aug 26 20:41:02 orangepi3b gnome-shell[2921]: Failed to post KMS update: drmModeAtomicCommit: Invalid argument Aug 26 20:41:02 orangepi3b kernel: [drm:vop2_plane_atomic_check] *ERROR* Unsupported linear format at Cluster0-win0 Aug 26 20:41:02 orangepi3b gnome-shell[2921]: Page flip discarded: drmModeAtomicCommit: Invalid argument Aug 26 20:41:02 orangepi3b gnome-shell[2921]: Failed to post KMS update: drmModeAtomicCommit: Invalid argument Aug 26 20:41:02 orangepi3b kernel: [drm:vop2_plane_atomic_check] *ERROR* Unsupported linear format at Cluster0-win0 Aug 26 20:41:02 orangepi3b gnome-shell[2921]: Page flip discarded: drmModeAtomicCommit: Invalid argument I strongly believe its some GNOME related error. Internet is full of this bug when people are using eGPUs. If anyone has any idea, please help
-
Struggle continues... Non-bootable images pushed me to look at u-boot stuff & my observation is this code block from default .csc file # Override family config for this board; let's avoid conditionals in family config. function post_family_config__orangepi3b_use_mainline_uboot() { display_alert "$BOARD" "mainline (Kwiboo's tree) u-boot overrides" "info" BOOTSOURCE="https://github.com/Kwiboo/u-boot-rockchip.git" BOOTBRANCH="branch:rk3xxx-2024.04" # specific commit tested is commit:ccef7dfc453bc03c5b5c61fa67d2745b96fa7da6 BOOTPATCHDIR="v2024.04-orangepi3b" # empty, patches are already in Kwiboo's branch:rk3xxx-2024.04 BOOTDIR="u-boot-${BOARD}" # do not share u-boot directory BOOTDELAY=1 # Wait for UART interrupt to enter UMS/RockUSB mode etc UBOOT_TARGET_MAP="BL31=${RKBIN_DIR}/${BL31_BLOB} ROCKCHIP_TPL=${RKBIN_DIR}/${DDR_BLOB};;u-boot-rockchip.bin u-boot-rockchip-spi.bin u-boot.itb idbloader.img idbloader-spi.img" unset uboot_custom_postprocess write_uboot_platform write_uboot_platform_mtd # disable stuff from rockchip64_common; we're using binman here which does all the work already # Just use the binman-provided u-boot-rockchip.bin, which is ready-to-go function write_uboot_platform() { dd if=${1}/u-boot-rockchip.bin of=${2} bs=32k seek=1 conv=fsync } # Smarter/faster/better to-spi writer using flashcp (hopefully with --partition), using the binman-provided 'u-boot-rockchip-spi.bin' function write_uboot_platform_mtd() { declare -a extra_opts_flashcp=("--verbose") if flashcp -h | grep -q -e '--partition'; then echo "Confirmed flashcp supports --partition -- read and write only changed blocks." >&2 extra_opts_flashcp+=("--partition") else echo "flashcp does not support --partition, will write full SPI flash blocks." >&2 fi flashcp "${extra_opts_flashcp[@]}" "${1}/u-boot-rockchip-spi.bin" /dev/mtd0 } } As previously stated the .csc file is extremely outdated & use old branches 2024.04 that do not include anything related to v2.1 for OPi3b, however the current branch do include (see attached image) so I changed in the above code references of 2024.04 to 2024.10 However this is one weird folder in the github which I cannot understanD & it contains a single .c file for OPi3b: https://github.com/Kwiboo/u-boot-rockchip/tree/rk3xxx-2024.10/board/xunlong/orangepi-3b-rk3566 v2.1 also used AP6256 Broadcom wifi/bt & I thus removed following functions / blacklist modules from .csc file, especially this sprd-* stuff. MODULES="sprdbt_tty sprdwl_ng" MODULES_BLACKLIST_LEGACY="bcmdhd" function post_family_tweaks_bsp__orangepi3b() { display_alert "$BOARD" "Installing sprd-bluetooth.service" "info" # Bluetooth on orangepi3b board is handled by a Spreadtrum (sprd) chip and requires # a custom hciattach_opi binary, plus a systemd service to run it at boot time install -m 755 $SRC/packages/bsp/rk3399/hciattach_opi $destination/usr/bin cp $SRC/packages/bsp/rk3399/sprd-bluetooth.service $destination/lib/systemd/system/ return 0 } function post_family_tweaks__orangepi3b_enable_services() { display_alert "$BOARD" "Enabling sprd-bluetooth.service" "info" chroot_sdcard systemctl enable sprd-bluetooth.service return 0 } Previously mentioned Step 2 was done additionally one file I placed here cd ~/build/cache/sources/rkbin-tools/firmware/wifi sudo wget https://raw.githubusercontent.com/orangepi-xunlong/firmware/master/nvram_ap6256.txt During build command I passed EXTRAWIFI=yes & BOARD_FIRMWARE_INSTALL=-full Working .csc is attached & I did get a bootable windows with working Broadcom wifi, It used module brcmfmac I can run everything using serial tty because hdmi was not working now. Will latter continue... but at least some success HDMI error Aug 26 05:17:05 orangepi3b kernel: dwhdmi-rockchip fe0a0000.hdmi: Detected HDMI TX controller v2.11a with HDCP (DWC HDMI 2.0 TX PHY) Aug 26 05:17:05 orangepi3b kernel: dwhdmi-rockchip fe0a0000.hdmi: registered DesignWare HDMI I2C bus driver Aug 26 05:17:05 orangepi3b kernel: dwhdmi-rockchip fe0a0000.hdmi: IRQ index 1 not found Aug 26 05:17:05 orangepi3b kernel: rockchip-drm display-subsystem: bound fe0a0000.hdmi (ops 0xffffc92fb9ba1078) Aug 26 05:17:05 orangepi3b kernel: input: hdmi_cec_key as /devices/platform/fe0a0000.hdmi/dw-hdmi-cec.1.auto/input/input0 Aug 26 05:17:05 orangepi3b kernel: #1: rockchip,hdmi Aug 26 05:17:06 orangepi3b systemd-logind[637]: Watching system buttons on /dev/input/event0 (hdmi_cec_key) Aug 26 05:17:11 orangepi3b kernel: dwhdmi-rockchip fe0a0000.hdmi: Rate 0 missing; compute N dynamically Aug 26 05:17:11 orangepi3b kernel: dwhdmi-rockchip fe0a0000.hdmi: Rate 0 missing; compute N dynamically Aug 26 19:41:27 orangepi3b kernel: dwhdmi-rockchip fe0a0000.hdmi: Rate 0 missing; compute N dynamically Aug 26 19:41:27 orangepi3b kernel: dwhdmi-rockchip fe0a0000.hdmi: Rate 0 missing; compute N dynamically I dearly request to community to regard OPi3b v2.1 as a completely different SBC with a different board entry in ./build orangepi3b.csc